The invention pertains to the field of knowledge management.
Data production is a function and by-product of the activities of most enterprises. However, while databases and data warehouses are effective for storage of data, their structures and capabilities generally reflect the needs of the enterprise""s discrete data production processes, rather than the needs of high level users whose function is to analyze the operation of the enterprise. Although a great deal of information about the enterprise is contained in its data, a high level analyst may not know whether desired information is actually reflected in any record that is possessed by the enterprise. Further, the user may not know where among an enterprise""s many electronic data storage facilities that information is likely to be found. The user is unlikely to know how to retrieve a desired record from a given database, since the specifications that are known to the user (i.e. the characteristics of the desired information) are unlikely to be the same as those used to identify a particular database record or data file that contains it. These problems are compounded when desired information is not reflected in an individual record, but instead can only be derived through synthesis of data contained in multiple records. In this case, the various records may be stored in different data storage facilities, and the manner of identifying and organizing records in each of the different data storage facilities is unlikely to be the same.
The invention addresses the above problems by providing an automated system, referred to as an xe2x80x9cinformation mart,xe2x80x9d for gathering data that is dispersed among the electronic data storage facilities of an enterprise, and for synthesizing and structuring the gathered data in accordance with the needs of user who require information to analyze the operation of the enterprise. The information mart provides a mechanism to organize the data in accordance with a predefined hierarchy of subjects for analysis, and a predefined set of relational attributes that are applied to all subjects of the subject hierarchy. The user is thereby provided with both hierarchical and relational access to the data. The user is enabled to browse the data in an intuitive manner through browser navigation tools provided in a graphical user interface that allow the user to provide input specifying a subject of interest and a relational attribute that characterizes the information desired about the subject. The browser further provides multiple simultaneous displays of content files containing related information. Multiple display of files provides juxtaposition of information that is derived from multiple sources and conveyed in multiple formats, thereby enhancing the user""s capacity for synthesizing and deriving further knowledge.
The information mart thus enables users to locate and browse the vast amount of disparate information that is used by an enterprise using an intuitive navigational metaphor that utilizes both a hierarchical and relational model. The information mart system described herein provides a mechanism to automate the production and simplify the navigation of pre-defined sets of related information that has been xe2x80x9charvestedxe2x80x9d from multiple, disparate electronic data sources that may be located both within and external to the enterprise.
Generic and preferred embodiments of systems and processes in accordance with the invention are described below, and the scope of the invention is defined in the appended claims. For purposes of summary, an information mart system in accordance with the invention generally has three aspects.
The first aspect of the information mart system is the logical design of the content file structure. The logical design is based upon a predefined hierarchy of subjects relating to an about which information may be desired. The logical design of the information mart further includes a predefined common set of xe2x80x9ctypes of informationxe2x80x9d, referred to herein as relational attributes, that are provided for each subject of the predefined hierarchy. Each subject within the subject hierarchy of the information mart is populated with displayable content files, and each content file is associated with one or more of the predefined relational attributes. Accordingly, every subject in the hierarchy effectively contains one or more groups of content files that are logically related because they each have the same relational attribute value. The logical design of the information mart can therefore be described as a group of content files that provide information pertaining to an enterprise, arranged hierarchically by subject, and arranged relationally by common relational attribute.
The second aspect of the information mart system is its physical implementation. The information mart content files are produced from data that resides in the various disparate electronic data storage facilities associated with the enterprise. A content file may be an existing file from the enterprise""s data storage facilities (e.g. an html document, a data table, a graphic, etc.), or it may be the result of processing of source data from a single source file (e.g. a graphic generated from a database record, a portion of a document, etc.), or it may be the result of processing of source data from multiple source files (e.g. a graphic generated using multiple database records from different databases). The content file may be stored locally in the information mart system, as would generally be true for a processed file, or it may be stored remotely and accessed as necessary, as might be the case for a file stored on a remote web server. Population of the information mart therefore involves, for every content file, determining the related source data for creating the content file, processing the source data, if necessary, and associating the physical address of the locally or remotely stored content file with the identifier of the content file.
To ensure that content files of the information mart system are current, the system may periodically update all locally stored content files. This may be accomplished by reacquiring and reprocessing all source data from the various electronic data storage facilities and rewriting the entire information mart. Alternatively, updating may be accomplished by comparing the time date stamp of each locally stored content file with the time date stamp of the current version of each file containing the content file""s source data, and selectively recreating out-of-date content files. As a further alternative, content files may be dynamically created (or updated) when requested for delivery to an end user, thus ensuring that the user receives current information.
Physical file organization may be accomplished through the use of a file folder tree structure that physically represents the hierarchy of subjects and the relational attributes corresponding to each subject. Alternatively, files may be virtually organized by means of a database that associates file identifiers with predefined subjects and predefined relational attributes.
The third aspect of the information mart system is browsing of content files by information users. A file server provides content files to a browser application for display to an end user in response to queries initiated by the end user through the browser. The queries are generated through selections by the user from among the predefined subject hierarchy and predefined relational attributes. The browser includes subject navigation tools for making these selections, including a subject navigation tool that allows the user to navigate through the subjects of the predefined subject hierarchy, and an attribute navigation tool that allows the user to navigate among the various predefined types of available information. The browser further includes multiple, simultaneously-displayed windows, each of which independently initiates queries to the file server in response to user manipulation of the navigation tools. Each window examines the record returned by the file server to determine its file type, and launches an appropriate application for displaying that file type in the window. Each window may further provide a pull-down list of alternate information mart files that correspond to the user specified subject and attribute but are not displayed, and the user may be allowed to initiate display of one of the listed alternate files.
The invention is considered to be embodied in an information mart system as disclosed herein, as well as in an information mart content creation facility, information mart server facility, and information mart browser facility as disclosed herein. The invention is further considered to be embodied in processes performed by the aforementioned facilities, as disclosed herein, and to be embodied in computer readable media storing control programming for configuring a data processing device to function as one of the aforementioned facilities disclosed herein, or to perform processes performed by the aforementioned facilities.
Further description of specific aspects of the invention and a preferred embodiment is provided below.